to-220 800+ V mosfet or BJT in linear mode at 60W

Nov 09, 2010 16 Replies

I'm looking for a mosfet or NPN that can handle 600Vds at 100mADC in linear mode within its SOA at 70 degrees C. It will be bolted to a substantial piece of aluminum to release the heat.



Most of the datasheets show the SOA at DC lower than this especially in to-220.



Unfortunately I'm stuck with to-220 since I replacing a part and I don't want to change the board if at all possible.



Also what is the lowest thermal resistance method to attach a to-220 and still maintain thermal isolation.



Grease and mica still the best?


Pretty much all FETs these days are targeted at switching apps, some even don't come with SOA data anymore. Check these out:

formatting link
formatting link

I'll leave that to others. As such high voltages, probably yes. I wouldn't risk it with silicone at 600V.

Regards, Joerg http://www.analogconsultants.com/ "gmail" domain blocked because of excessive spam. Use another domain or send PM.

Is Jeorg recommending toobs? I wonder, will a pair of 6L6GCs fit in the space required? ;-)

Tim

Deep Friar: a very philosophical monk. Website: http://webpages.charter.net/dawill/tmoranwms

I am partial to 6KD6 or PL509/519. Yep, even when building tube amps I was a cheapskate :-)

Regards, Joerg http://www.analogconsultants.com/ "gmail" domain blocked because of excessive spam. Use another domain or send PM.

The audiophools and TV restorers picked up on both of those. Most sweeps are pricey now!

Tim

Deep Friar: a very philosophical monk. Website: http://webpages.charter.net/dawill/tmoranwms

So did us RF guys. I built a kilowatt amp out of five PL509. In Germany ham radio had a loophole, 150W rated plate dissipation for the highest license class. So ... turns out the PL509 was officially 30W but could do a lot more. My amp sustained 1200W output on shortwave. Here it is:

formatting link

That law later changed, to 750W output no matter what the plate rating was. So I built this one and it could deliver 750W almost into a dead short if needed:

formatting link

Here is the datasheet of the two tubes in the back of it, the filament alone burns 100W, each:

formatting link

Yep. But power FETs have become very economical. Those were totally out of league for teenagers in the 70's. I think I owned a total of two power FETs back then, in TO-3 package. They moved from project to project.

Regards, Joerg http://www.analogconsultants.com/ "gmail" domain blocked because of excessive spam. Use another domain or send PM.

While tubes might be interesting in some applications, the quite limited filament to cathode voltage limit makes it hard to design any "interesting" systems without a huge number of filament windings on the mains transformer.

Maybe bolt it to a heat spreader, then attach the spreader to the sink through mica -- if you can make it fit.

Tim Wescott Wescott Design Services http://www.wescottdesign.com Do you need to implement control loops in software? "Applied Control Theory for Embedded Systems" was written for you. See details at http://www.wescottdesign.com/actfes/actfes.html

"mook johnson" wrote in news:no1Co.8942$ snipped-for-privacy@newsfe03.iad:

Check IXYS they have some Mosfets for linear operation upto 1500V.Here is the selection chart.I know Newark and digikey stock some I dont know about the 1200-1500V ones though. You may be able to get by with a

1000V one I just had time to look at one Datasheet they have a couple.

Had to shorten the url

formatting link

I remember being shown photographs of a British AM, 400kW, broadcast transmitting station in the 1970s. One showed a room with a row of M-G sets, similar in size to those you'd see in a plating shop.

"What the hell are those for?"

"They're the filament supplies for the finals."

A couple of volts, at hundreds of amps, IIRC

"For a successful technology, reality must take precedence over public relations, for nature cannot be fooled." (Richard Feynman)

Not sure what you mean by that but the PL509 was designed for series heating. Those tubes can tolerate mains voltage between filament and cathode. The QB5/1750 tube from my big amps have direct-heat cathodes, IOW the cathode is the filament.

Regards, Joerg http://www.analogconsultants.com/ "gmail" domain blocked because of excessive spam. Use another domain or send PM.

Joerg Inscribed thus:

Hey I've got a couple of those ! Never used them though.

Best Regards: Baron.

Nice. Your construction makes me drool! :-)

Ed

P-series (300 mA) tubes typically had about 200 V maximum filament/cathode voltage limit, thus, the filament string could be connected directly to the 220 V mains trough a resistor (to set the

300 mA loop current) and a VDR (to reduce the startup current shock), but this assumed that the cathode was close to ground (chassis voltage).

However, if you wanted to use a triode or pentode for series stabilized +400 V supply, the cathode is at +400 V and typically require a separate filament winding. The same is true for the upper tube(s) in totem pole or H-bridge configurations.

Next time I do some serious tubifying, I'm going to build a cheezy AC source. I've got some lovely ferrite toroids that are three inches across. I figure, put 100kHz into the primary, and I can get 3 or 6V per turn, enough that I only need to loop hookup wire once or twice through the core to heat up a standard tube. I could even have half the toroid razzing out the front of a chassis, so I can loop wires from anything, just as test equipment.

Such will definitely come in handy for a class D tube amplifier, with antiparallel diodes and plate-screen ("baker") clamp diodes.

Tim

Deep Friar: a very philosophical monk. Website: http://webpages.charter.net/dawill/tmoranwms

Must be more than 200V because the peak in a 230V AC circuit is 325V. As a kid I repaired and parted out lots of old tube TV sets. Some of the more luxurious sets (with motor tuner and stuff like that) were so full of tubes that the string used up almost the whole 230V, very little resistance in series with it.

[...]
Regards, Joerg http://www.analogconsultants.com/ "gmail" domain blocked because of excessive spam. Use another domain or send PM.

My reference is the blue cover 1964 Telefunken "Taschenbuch" .

Except for the UK, 220 V was the mains voltage in Europe.

150-200 V was handled by the filament and the series resistance handled the rest.

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required